* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download In vivo quantification of peroxisome tethering to chloroplasts in
Survey
Document related concepts
Transcript
Plant Physiology Preview. Published on October 30, 2015, as DOI:10.1104/pp.15.01529 1 2 Running Title: Peroxisome-chloroplast tethering 3 *Corresponding author: 4 Imogen Sparkes 5 Biosciences, 6 University of Exeter, 7 Geoffrey Pope Building, 8 Stocker Road, 9 Exeter, 10 EX4 4QD, 11 UK 12 [email protected] 13 Tel: +44 (0)1392 723454 14 1 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Copyright 2015 by the American Society of Plant Biologists 15 Title: In vivo quantification of peroxisome tethering to chloroplasts in tobacco 16 epidermal cells using optical tweezers 17 18 Authors: H. Gao1, J. Metz1#, N. A. Teanby2#, A. D. Ward3, S. W. Botchway3, B. 19 Coles3, M. R. Pollard3, I. Sparkes1* 20 1 21 Affiliations: 22 Road, Exeter, EX4 4QD, UK 23 2 24 Queen's Road, Clifton, Bristol, BS8 1RJ, UK 25 3 26 Complex at Harwell, Didcot, Oxon OX11 0FA, UK Biosciences, University of Exeter, Geoffrey Pope Building, Stocker School of Earth Sciences, University of Bristol, Wills Memorial Building, Central Laser Facility, Science and Technology Facilities Council, Research 27 28 # 29 *corresponding author Authors contributed equally 30 31 One sentence summary: Optical tweezers shows that peroxisomes are strongly 32 tethered to chloroplasts, and more weakly to other unknown structures. 33 Footnotes: 34 IS and HBG are funded by the BBSRC (BB/I006184/2) and an STFC user access 35 facility grant (LSP907), JM is funded by a Wellcome Trust Institutional Strategic 36 Support Award (WT097835MF). 37 38 M.R. Pollard’s current address, DFM A/S, K. Lyngby, Denmark 2800 39 40 [email protected] 41 2 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 42 Abstract 43 Peroxisomes are highly motile organelles that display a range of motions within a 44 short time frame. In static snapshots they can be juxtaposed to chloroplasts which has 45 led to the hypothesis that they are physically interacting. Here, using optical tweezers 46 we have tested the dynamic physical interaction in vivo. Using near-infrared optical 47 tweezers, combined with TIRF microscopy, we were able to trap peroxisomes and 48 approximate the forces involved in chloroplast association in vivo, and observed 49 weaker tethering to additional unknown structures within the cell. We show that 50 chloroplasts and peroxisomes are physically tethered through peroxules, a poorly 51 described structure in plant cells. We suggest peroxules have a novel role in 52 maintaining peroxisome-organelle interactions in the dynamic environment. This 53 could be important for fatty acid mobilisation and photorespiration through interaction 54 with oil bodies and chloroplasts, highlighting a fundamentally important role for 55 organelle interactions for essential biochemistry and physiological processes. 56 57 58 Introduction 59 60 A combination of genetically encoded fluorescent probes, advances in light 61 microscopy and interdisciplinary approaches have revolutionised our understanding of 62 organelle transport. Organelle movement in highly vacuolated leaf epidermal cells 63 appears erratic with individual organelles undergoing a range of movements within a 64 relatively short time frame; stop-go, change direction (trajectory) and can move at 65 varying speeds. Use of pharmacological inhibitors indicated a role for actin, and 66 therefore myosins in this process, however myosin-organelle specificity is poorly 67 characterised (Buchnik et al., 2015; Madison and Nebenführ 2013; Tamura et 68 al.,2013). We are therefore still at a relatively rudimentary stage in the understanding 69 of the molecular and physical control, and interaction of, organelles in plant cells 70 compared with that known in other model systems (Prinz 2014; Hammer and Sellers 71 2012). However, it is clear that organelle movement plays important roles in 72 physiological processes in plants; reduced movement effects growth and 73 development, and movement is correlated with responses to extracellular stresses such 74 as pathogens and heavy metals (see references in Madison and Nebenführ 2013, 75 Buchnik et al., 2015 and Sparkes 2011). Organelle interactions in other systems have 3 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 76 important roles in calcium and lipid exchange setting a precedent for physiologically 77 important roles in plants (Prinz, 2014). However, characterisation of the molecular 78 factors required to physically tether organelles as opposed to those which function in 79 the exchange of molecules at the interaction site is challenging. Monitoring organelle 80 interactions in highly vacuolated plant epidermal cells are further complicated by the 81 constraints imposed by the large central vacuole. Static snapshots provided through 82 electron microscopy of highly vacuolated cells, where the vacuole can effectively 83 ‘push’ organelles together giving the impression of direct interaction between 84 organelles, is not a suitable method to determine dynamic interactions. Other 85 techniques such as the laser induced shockwave by explosion method used by Oikawa 86 et al., (2015) works globally without directly manipulating the individual organelle. 87 Here, using optical tweezers with sub-micron precision, we provide a means to assess 88 and quantify the dynamic interaction between peroxisomes and chloroplasts in vivo in 89 leaf epidermal cells. 90 91 92 Peroxisomes are responsible for several biochemical reactions including the 93 glyoxylate cycle and β-oxidation which provides an energy source for germination in 94 oilseeds. They also produce and scavenge free radicals, synthesise jasmonic acid (JA), 95 IAA and are required for photorespiration (see references in Hu et al. 2012). The 96 photorespiratory pathway spans peroxisomes, chloroplasts and mitochondria where 97 phosphoglycolate produced in the chloroplast is converted back to 3-P-glycerate. It 98 has been suggested that functional connectivity between these organelles accounts for 99 the close association observed in ultrastructural micrographs (Fredericks and 100 Newcomb, 1969). Several Arabidopsis pex10 (peroxisomal membrane protein) 101 mutants show altered chloroplast-peroxisome juxtaposition with a defect in 102 photorespiration while others do not (Schumann et al., 2007; Prestele et al., 2010). 103 Both Clumped Chloroplasts 1 (CLMP1) and Chloroplast Unusual Positioning 1 104 (CHUP1) encode for proteins that localise to the chloroplast, with CHUP1 playing a 105 role in cp-actin formation (Yang et al., 2011; Oikawa et al., 2003, 2008; Schmidt von 106 Braun & Schleiff 2008). Whilst CHUP1 and CLMP1 affect chloroplast positioning, 107 they have differential effects on peroxisome and mitochondrial location; clmp1 causes 108 chloroplast clustering without affecting mitochondria or peroxisome location (Yang et 4 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 109 al., 2011), whereas chup1 was reported to affect peroxisome location (Oikawa et al., 110 2003). In vitro analysis through density centrifugation highlighted chloroplast 111 sedimentation with peroxisomes under certain conditions (Schnarrenberger and 112 Burkhard, 1977), although this does not necessarily reflect organelle interaction in 113 live cells. Peroxisome proteomics studies have been hampered by difficulties in 114 isolating pure peroxisomal fractions (Bussell et al., 2013). This could be indicative of 115 interaction, where associated membranes are isolated together, or ‘sticky’ non-specific 116 contaminating chloroplast membranes. The work by Oikawa et al. (2015) provides 117 insight into the physiological processes controlling peroxisome-chloroplast interaction 118 (photosynthetic dependent), but they did not determine the effective baseline force 119 required to move peroxisomes that were not next to chloroplasts under control or 120 altered environmental conditions. Comparisons between the relative forces required to 121 move peroxisomes next to chloroplasts versus those which are not next to chloroplasts 122 are critical in understanding and probing the physical interaction between the two 123 organelles; hypothesis being that tethering would increase the force required to move 124 peroxisomes compared to organelles which are not tethered. Since peroxisomes have 125 diverse biochemical roles that affect a wide range of physiological processes 126 throughout the plant life cycle (Hu et al. 2012), then an understanding of if and how 127 peroxisomes may interact with other subcellular structures is likely to be an important 128 consideration for efficient peroxisome function. 129 130 131 132 133 Peroxisomes are highly pleomorphic, dynamic organelles bounded by a single 134 membrane (Hu et al., 2012), whose movement is driven by acto-myosin dependent 135 processes (Jedd and Chua, 2002; Mano et al., 2002; Mathur et al., 2002; Avisar et al., 136 2008; Sparkes et al., 2008). Tubular emanations termed peroxules (Scott et al., 2007) 137 can extend from the main peroxisome body, yet it is unclear what function they may 138 play. Formation is quite frequent in hypocotyl cells (Sinclair et al., 2009; Mano et al., 139 2002; Cutler et al., 2000), can occur around chloroplasts in cotyledonary leaf 140 pavement cells (Sinclair et al., 2009), and do not always form from the trailing edge 141 of the peroxisome (Sinclair et al., 2009). Exogenous addition of hydroxyl ROS, or 142 exposure to UV light, induces peroxule formation (Sinclair et al., 2009). It has been 5 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 143 suggested that they represent increased surface area for increased biochemical 144 function, or might represent a morphological precursor for peroxisome division (Jedd 145 and Chua 2002). Based on subcellular co-alignment, a retro-flow model for potential 146 exchange of luminal content between the ER and peroxisome through the peroxule 147 has been suggested (Sinclair et al., 2009; Barton et al., 2013). However, these studies, 148 as with many others, interpret the close association between organelles to indicate 149 physical connectivity between organelles, whereas, in fact, in highly vacuolated leaf 150 epidermal cells organelles can be closely packed within the cytoplasm due to mere 151 spatial constrictions generated through the large central vacuole. This is further 152 complicated by the highly motile, and seemingly stochastic nature of acto-myosin 153 driven organelle movement resulting in frequent apparent organelle ‘collisions’ which 154 may not reflect a functional requirement for organelle interaction. 155 156 Optical trapping provides a highly specific and sensitive means to measure physical 157 connectivity between organelles. By focussing an infrared beam, it allows the user to 158 trap objects which have a significantly different refractive index to the surrounding 159 media. Upon trapping, the user can then move the trapped object relative to its 160 original position to gain an understanding of whether the movement affects the 161 position and motion of other structures (such as other organelles) which may be 162 physically attached to the trapped organelle. For example, unlike the ER, Golgi bodies 163 are amenable to trapping. By trapping and micromanipulating (i.e. precisely moving) 164 the Golgi, a physical association between the ER and Golgi was determined in a 165 qualitative manner (Sparkes et al. 2009b). Here, we have developed a system to 166 generate quantitative measures for organelle interaction by standardising and 167 automating how far we move the trapped organelle (which we call the translation 168 step) at a defined speed, and assessing how trapping efficiency alters in response to 169 the power of the laser trap itself. By using these parameters we can then model the 170 forces imparted on the organelle providing further insight into the tethering processes. 171 Our results indicate that peroxisomes are amenable to being trapped, that they 172 physically interact with chloroplasts in leaf epidermal cells, and surprisingly that 173 peroxisomes are also tethered to other unknown structures within the cell. This 174 approach therefore highlights that organelle interactions within plant cells are not 175 random but regulated through tethering. In addition we provide a novel role for 6 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 176 peroxules, and a simple biophysical model to describe peroxisome motion during the 177 trapping process. 178 179 7 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 180 Results 181 182 Peroxisome association with chloroplasts is specific 183 184 For organelles to interact they must move and physically ‘sit’ or reside next to one 185 another in a coordinated manner. To determine how peroxisomes move relative to 186 chloroplasts we observed both peroxisomes, chloroplasts and Golgi bodies within the 187 same tobacco leaf epidermal cell and assessed how long either peroxisomes or Golgi 188 resided next to chloroplasts. As organelles are physically constrained by the large 189 central vacuole and can be ‘pushed’ together, Golgi were monitored as they are not 190 functionally related to chloroplasts and so act as an inherent control. We observed that 191 the average residency time of peroxisomes was significantly higher than that of Golgi 192 bodies on chloroplasts; 1.46 +/- 0.35 minutes (n=17) and 0.42 +/- 0.05 minutes (n=51) 193 respectively t-test p<0.001 (Supp movie 1). Due to these observations, and functional 194 connectivity through the photorespiratory pathway, we investigated whether 195 peroxisomes physically interact with chloroplasts in vivo. 196 197 Peroxisomes are associated with chloroplasts in an actin independent manner 198 In a motile system it is difficult to discriminate between physical tethering processes 199 between two organelles from acto-myosin driven events. We therefore assessed 200 whether interaction characteristics were actin dependent in the first instance. Note, the 201 concentration of latrunculin b used is sufficient to depolymerise actin and cause 202 cessation of organelle movement (Sparkes et al., 2009a; Sparkes et al., 2008). 203 204 The average percentage of chloroplasts with a juxtaposed peroxisome in the presence 205 or absence of actin (latrunculin b treated) were not significantly different from one 206 another; 22 +/- 5% and 23 +/- 3% respectively, t-test p>0.8, data taken from 20 images 207 covering 0.4mm2 leaf epidermal area. Using optical tweezers we then tested whether 208 these results indicated a peroxisome-chloroplast tethering mechanism in both motile 209 and non-motile (latrunculin b treated) samples. By trapping and subsequently moving 210 the peroxisome within the cell (Fig 1A-D) we observed that upon turning the trap off 211 the peroxisome recoiled back towards its place of origin irrespective of chloroplast 212 presence (Supp Movie 2B,C,D). This process has not been previously observed using 213 other techniques. On several occasions peroxules were observed from the trailing 8 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 214 edge of the peroxisome (Supp Movie 2B,D). Upon actin depolymerisation, trapped 215 peroxisomes displayed similar characteristics; peroxule formation and peroxisome 216 recoil upon turning the trap off (Fig 1E-H, Supp movie 3A,B). These results indicated 9 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 217 that peroxisomes are tethered to chloroplasts and unknown structures in the cell, and 218 that peroxules may represent the site of tethering. 219 220 To test the hypothesis that peroxisomes are tethered to chloroplasts we set about 221 quantifying whether the average laser power required to trap and move peroxisomes 222 was dependent on chloroplast positioning and / or actin. The rationale here is that 223 trapping efficiency and movement are dependent on optical trap strength where 224 tethering, which acts as an opposing force, would impede the movement of the 225 trapped organelle causing it to escape the trap. Trapping refers to an organelle which 226 can be trapped and remains in the trap over the 6μm translation distance (Fig 1). Of 227 the fifty organelles from independent cells that underwent the trapping routine (which 228 constituted 5 samples of 10 organelles) there was a clear trend that increasing optical 229 laser power (from 24 to 50mW) resulted in an increase in the number of trapped 230 peroxisomes (20-38% increase) irrespective of actin or chloroplast association. 231 However, peroxisomes which were next to chloroplasts were harder to trap. 232 Significantly fewer chloroplast associated (cp) peroxisomes were trapped when 233 compared to non chloroplast (non-cp) associated peroxisomes under either motile or 234 immotile (latrunculin b treated) conditions at a given laser power; 50mW optical 235 trapping laser power resulted in average trapping of 38 +/- 2% cp and 56 +/- 7% non-cp 236 in the motile system, and 36 +/- 4% cp and 70 +/- 4% non-cp in the immotile system, t- 237 test p<0.05 comparing cp to non-cp under a given condition. These results indicate 238 that peroxisomes are tethered to chloroplasts and that this phenomenon is independent 239 of actin. The trapping efficiency of non-cp peroxisomes in the motile system 240 compared to the immotile system was significantly reduced (t-test p<0.15) and could 241 be due to a number of reasons; trapped peroxisome being knocked out of the trap by 242 passing organelles, docking the peroxisome onto actin filaments during the translation 243 or moving a trapped organelle into a cytoplasmic stream (Supp movie 2). 244 245 Peroxisome tethering can also be quantified by monitoring the recoil of the 246 peroxisome back towards its origin after turning the trap off (Fig 1, termed recovery 247 displacement). However, inherent difficulties of organelles escaping the trap and 248 responding to the acto-myosin driven elements after turning the trap off reduced the 249 number of organelles that could be assessed in this manner; for cp motile, and cp / 10 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 250 non-cp non motile system (latrunculin b treated) between 66-84% were measurable 251 compared with 21% for non-cp motile system. Observations of the small number of 252 organelles which showed recoil back towards the trap origin (recovery displacement, 253 Fig 1), rather than movement in an opposite direction in the motile system, indicated 254 that recoil was significantly larger for cp compared to non-cp in both motile and non- 255 motile conditions: for the motile system cp recovery displacement 3.92 +/- 0.36μm 256 (n=16) compared with non-cp recovery displacement 2.50 +/- 0.29 μm (n=6) p<0.007; 257 for the non-motile system cp recovery displacement 3.65 +/- 0.45μm (n=14) compared 258 with non-cp recovery displacement 1.22 +/- 0.20μm (n=23) p<0.001. All data were 259 taken using 50mW trapping laser power with a 5.3 second recovery period. 260 261 Quantifying the peroxisome-chloroplast tethering process: a novel role for 262 peroxules 263 264 The above observations clearly indicate that peroxisomes are tethered to chloroplasts, 265 and that this phenomenon is independent of actin. To further characterise the effects 266 of tethering, the opposing forces generated by the acto-myosin component were 267 removed from the system (latrunculin b treatment). Here, we assessed (1) the 268 relationship between peroxisome behaviour in the trap and trapping laser power over 269 a larger range of laser powers, and (2) behaviour of displaced peroxisomes after 270 turning the trap off (Fig.1,2). All of these observations were carried out under 271 latrunculin b treatment so that any interactions are due to tethering and not the acto- 272 myosin system. 273 274 Peroxisomes were either trapped, not trapped or escaped the trap during translation 275 (Fig 2c; movie S3C-E). As expected, the trapping laser power correlated with the 276 observed percentage trapping for both cp and non-cp peroxisomes (Fig.2A, B). 277 However, at laser powers of 37 mW and above there was a significant difference 278 between the trapping of cp and non-cp peroxisomes, with cp peroxisomes escaping 279 the trap more readily and non-cp peroxisomes being trapped and remaining in the trap 280 over the 6μm translation (Fig 2, Supp Fig 1). Taken together this is indicative of more 281 force being required to trap and move peroxisomes away from chloroplasts. 282 Additionally, upon turning the trap off, cp trapped peroxisomes underwent a 11 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 283 significantly larger recovery displacement (i.e. recoil, Fig 1D) than non-cp trapped 284 organelles: cp recovery displacement 4.39+/-0.17μm (n=94), non-cp recovery 285 displacement 2.93+/-0.17μm (n=91) using a laser power of 37mW, which has a t test 286 probability value of p<0.001. This proves that peroxisomes are tethered to 287 chloroplasts in vivo in tobacco leaf epidermal cells. The above data were generated 288 under a long recovery period (21.5 seconds rather than 5.3 seconds) to allow 289 organelles to reach their equilibrium position, which improves the accuracy of the 290 force determination discussed later. 291 292 293 Peroxule formation can occur upon exposure to the trapping laser prior to and during 294 the translation, however the frequency of formation is independent of the power of the 295 optical trapping laser indicating that formation is not solely due to exposure to the 296 trapping laser (Supp. table 1). Interestingly, both cp (38% n=170) and non-cp (37% 12 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 297 n=183) peroxisomes had a similar propensity to form peroxules, but the relative 298 percentages between formation in response to exposure to optical trap versus 299 translation differed (Supp table 1). It is unclear why more peroxules would form in the 300 absence of chloroplast positioning prior to translation (2.9% compared with 12%), but 301 we speculate that formation may occur in response to stress which is ameliorated by 302 the antioxidant properties of the chloroplast (Sinclair et al., 2009; Asada 2006), or 303 non-cp peroxisomes are tethered to structures whose positioning alters in response to 304 trapping the peroxisome. 305 306 During peroxisome translation peroxule formation is more frequent in cp versus non- 307 cp associated peroxisomes (28.2% compared with 15.3%, table S1), correlative with 308 peroxules being the visible manifestation of the tether to chloroplasts. The tip (point 309 of origin; see Fig 1G asterisk) of peroxules moved less during the translation process 310 in cp versus non-cp association, indicative of an anchored tether; cp 1.85+/-0.3μm 311 (n=37), non-cp 2.35+/-0.2μm (n=46). In comparison, during the recovery period 312 peroxule tip displacement was much smaller, with values for cp and non-cp being 313 similar; cp 0.89 +/- 0.2μm, 314 peroxule tip; see Fig 1G asterisk) is anchored, one would expect a higher level of 315 peroxisome movement (i.e. recoil) during the recovery period to correspond with a 316 lower level of peroxule tip movement during the translation; unlike non-cp samples, 317 there is a cluster of cp samples indicative of such behaviour suggesting strongly 318 anchored tether bases (Fig. 3). non-cp 1.13+/-0.1μm. If the base of the tether (i.e. 319 320 Biophysical modelling of peroxisome recoil indicates differences in relative forces 321 for peroxisome interactions 322 323 Since both cp and non-cp peroxisomes exhibit different trapping (Fig. 2) and recovery 324 (Fig. 3) behaviours, we sought to understand the forces involved in this process; 325 specifically is it only the recoil distance which changes or are there changes in the 326 tether properties between cp and non-cp peroxisomes? To allow us to distinguish 327 between tether properties and changes in recoil distance we used a simple viscously 328 damped spring model to estimate the tether stiffness (i.e. spring constant) and tether 329 tension forces (i.e. initial recovery force) involved in the recovery process (Fig 4, 13 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 330 Supp. note, Supp Fig 2-4). This first approximation indicates that tether stiffness 331 values are similar for non-cp and cp (Fig 4B,C) and that differences in the recovery 332 forces are solely due to the more rigid anchoring of cp associated peroxisome tethers, 333 which leads to greater tether extension and subsequently greater recovery 14 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 334 displacement and initial recovery forces (Fig 4). In other words the biological 335 structure that forms the tether between cp and non-cp peroxisomes behaves in a 336 similar manner (i.e. similar stiffness), but the base of the tether (i.e. anchor point) 337 moves less for cp peroxisomes thus generating more tension during translation and 338 resulting in greater recoil. Here, non-cp peroxisomes are tethered to a structure which 339 has greater mobility than chloroplasts during the trapping routine, so that upon 340 moving non-cp peroxisomes, the tethered structure is also able to move to a certain 341 extent resulting in lower tension ‘build up’ during the translation process. As we 342 cannot independently estimate cytoplasmic viscosity in our system, this approach can 343 only be used to determine relative differences in forces between cp and non-cp 344 associated peroxisomes. However, using a reasonable value of 0.06 Pa s (Scherp and 345 Hasenstein 2007) gives tether stiffnesses of ~1pN/μm and initial recovery forces of 346 ~1-4pN (median values from Fig 4). 347 348 349 350 15 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 351 Discussion 352 353 By using optical tweezers we clearly show that peroxisomes can be tethered to 354 chloroplasts, and that relative differences in tethering strength highlight additional 355 subcellular interactions. Moreover, these tethers can be observed in several instances 356 as peroxules (Supp movie 2, 3). Such tethers are not solely restricted to chloroplast 357 interaction, but are also prevalent on non-cp peroxisomes (Supp movie 2, 3). In the 358 latter case, the tether interaction is either unstable or the structure it is tethered to is 359 more readily motile, accounting for the movement of the peroxule tip base during 360 translation. The mechanism of peroxule formation and extension is unclear, but the 361 rapid rate of extension makes de novo synthesis unlikely. Alternatives could be that 362 the bounding membrane itself is deformable, or that peroxules are tightly coiled 363 around the peroxisome and indistinguishable from the fluorescence signal arising 364 from the lumen of the main peroxisome body. It is unclear if the connectivity between 365 peroxisomes and chloroplasts is direct or indirect as positioning could be mediated 366 through interaction with the ER. The ER forms a basket around the chloroplasts 367 (Schattat et al., 2011), and in vitro optical trapping data inferred a chloroplast-ER 368 connection in Arabidopsis and pea leaf cells (Andersson et al., 2007). Peroxisomal 369 membrane protein Pex3p has been implicated in acting as a direct tether between the 370 ER and peroxisomes in S. cerevisiae (Knoblach et al., 2013). However, the complex 371 biogenetic link between peroxisomes and the ER has been, and continues to be, 372 debated within the community (Hu et al., 2012). Our previous observations of ER 373 responses upon trapping and moving Golgi highlight that a large percentage of the ER 374 is freely mobile, however chloroplast-ER interactions were not investigated (Sparkes 375 et al., 2009b). Therefore, if chloroplast-peroxisome connectivity is mediated by an ER 376 bridge, then perhaps the ER is highly constrained around chloroplasts which could 377 lead to greater recoil of trapped cp peroxisomes compared with non-cp cases. This is 378 an area of future study requiring further development of the imaging system. Using 379 the approaches developed here, future studies will enable the molecular and 380 physiological consequences of peroxisome-organelle interaction to be studied, and 381 could also be used to study the formation of membrane extensions. 382 383 Interactions between organelles are likely required for communication and transport. 384 Examples in yeast and mammals infer a requirement for lipid and calcium exchange 16 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 385 (Prinz 2014). In plants reports for ER-Golgi (Sparkes et al., 2009b), nucleus-plastid 386 (Higa et al., 2014), ER-chloroplast (Andersson et al., 2007; Mehrshahi et al., 2013), 387 peroxisome-oil body (Thazar-Poulot et al., 2015) interactions have been made, along 388 with a recent report from Oikawa et al. (2015) inferring a chloroplast-peroxisome 389 interaction in Arabidopsis mesophyll cells. This study, along with previous reports, 390 indicates peroxisomes undergo light dependent morphological changes (Desai and Hu 391 2008;Oikawa et al., 2015). Furthermore, by effectively inducing a localised 392 intracellular shock wave, Oikawa et al. inferred light, and photosynthesis, dependent 393 connections between peroxisomes and chloroplasts. Here, using a complementary 394 approach we trap individual peroxisomes in tobacco leaf epidermal cells, and 395 additionally compare the responses between chloroplast associated and non-associated 396 peroxisomes. Our results provide a clear indication of interaction of peroxisomes with 397 chloroplasts, and other unknown structures, and we provide a biophysical model for 398 the forces involved in the tethering process. We have also visualised the tethering 399 process through peroxule production, observations which were not made in the work 400 of Oikawa et al. and therefore suggest a novel role for peroxules in maintaining 401 physical connectivity between peroxisomes and the structure(s) to which it is tethered 402 to. The two techniques infer forces for the peroxisome-chloroplast interaction, but by 403 the very nature of the techniques the forces relate to different biological aspects of the 404 interaction; Oikawa et al. models the force required to push the two organelles apart 405 (23-61 fN nm-2), whereas here we model the forces imparted on the organelle after 406 they have been separated. It is important to note that the speed used to separate the 407 organelles using optical tweezers is within the range of reported peroxisome speeds in 408 an unperturbed system (Sparkes et al. 2008), and so cytoplasmic viscosity will affect 409 interactions in a way in which reflects the native motile system. Whereas the force 410 imparted on peroxisomes using the focused femtosecond laser technique was reported 411 to be so large that the effects of cytoplasmic viscosity would not hinder free 412 peroxisome motion, and are therefore negligible in their system. We do not infer a 413 precise force for trapping and moving the organelle (as viscosity values are currently 414 unknown for the system) and so compare the trapping profiles of chloroplast 415 associated and non-associated peroxisomes in response to the trapping laser power. 416 Both systems therefore provide different force components and have different 417 strengths and weaknesses in assessing the peroxisome-chloroplast interaction. 418 Furthermore, the basic spring model provides a baseline for interactions and will be 17 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 419 useful in testing how effective tension and stiffness change under altered 420 environmental conditions that may regulate the interaction between peroxisomes and 421 chloroplasts. For example, as photorespiration and photosynthesis may affect 422 interaction, does the rate of recoil of a trapped peroxisome change indicating a 423 ‘tighter’ tethering process between peroxisomes and chloroplasts or other structures 424 within the cell, and does this altered response affect tether stiffness rather than 425 tension? 426 427 Several organelles in plant cells produce tubular emanations; stromules, matrixules 428 and peroxules extend from chloroplasts, mitochondria and peroxisomes respectively 429 (Scott et al., 2007; Mathur et al., 2012; Hanson and Sattarzadeh 2013). Mapping 430 stromule dynamics, and the movement of protein and small molecules lend support to 431 a role in communication. However, contradictory data from different groups on 432 molecular exchange between stromules makes this an interesting and contentious area 433 of research (Hanson and Sattarzadeh 2013; Mathur et al., 2013). Here, our results 434 infer a similar role in communication, and we propose that tubular emanations are a 435 consequence of organelles attempting to maintain connections in the highly dynamic 436 intracellular environment. Peroxule formation occurs in response to hydroxyl reactive 437 oxygen species (ROS) with a concomitant reduction in peroxisome speed (Sinclair et 438 al., 2009). This could be interpreted as a response to maintain connections between 439 peroxisomes and another organelle whose motility has not been affected, or has been 440 increased during this treatment, effectively increasing the spatial separation between 441 the two organelles. Whilst the biophysical model provided herein reveals pN force 442 measurements imparted on the organelle during the recovery process, it also gives an 443 indication of the force required to pull the organelle micron distances. Here, the motor 444 force to separate organelles is expected to be the same or greater than the force 445 required for the organelles to be ‘pulled’ back towards their resting position (i.e. 446 referred to as the restorative force in the biophysical model). This approach could 447 therefore determine how motor regulation is controlled in order to maintain 448 peroxisome movement under conditions where interaction with chloroplasts is up / 449 down regulated. 450 451 Organelle movement plays important roles in growth, development and in response to 452 (a)biotic stresses (see references in Madison and Nebenführ 2013, and Buchnik et al., 18 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 453 2015). In a wider context, the results presented herein allow us to start to bridge the 454 interface between organelle movement and interaction, and the forces involved in 455 these processes. Whist future studies are required to validate the force measurements 456 with known cellular viscosities, in broader terms, these studies demonstrate that 457 interactions between organelles such as peroxisomes and chloroplasts in plant cells 458 are not random, but are controlled through tethering mechanisms which can be 459 quantified using optical tweezers. Regulation of organelle interaction / association 460 will be controlled by motor driven movement to position organelles next to one 461 another to allow tethering processes to occur. The force balance between these two 462 processes therefore needs to be viewed in conjunction to describe organelle motion 463 and positioning. Organelle interactions in plants could be required for communication 464 and so future studies pinpointing the tethering and motor components could provide a 465 novel way in which to control subcellular communication. 466 467 468 469 An interdisciplinary approach will be needed to fully characterise the molecular and 470 physiological role(s) of peroxisome-chloroplast interactions, and interactions with 471 other unknown organelles which could include lipid bodies. Current evidence points 472 towards photosynthetic dependent processes and a role for PEX10 in peroxisome- 473 chloroplast interactions (Oikawa et al., 2015; Schumann et al., 2007; Prestele et al., 474 2010). It will also be interesting to assess what role ROS signalling may play in these 475 interactions (Sandalio and Romero-Puertas, 2015), and whether the exchange of 476 additional small molecules such as IAA and JA may be facilitated by organelle 477 interaction. Future genetic screens and proteomic approaches will pinpoint the 478 complex of proteins necessary for interaction. The essential domains required for 479 tethering will be mapped using biophysical means, such as optical tweezers, to 480 quantify effects on peroxisome-chloroplast / organelle interaction. Ultimately, the 481 analysis of resulting lines deficient in the tethering process will provide both 482 molecular, biochemical and physiological evidence for the role of peroxisome- 483 chloroplast / organelle interaction. 484 485 486 19 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 487 Methods 488 489 Plant material and sample generation 490 Nicotiana tabacum plants were grown and transiently transformed according to 491 Sparkes et al. (2006) GFP-SKL (Sparkes et al., 2003), YFPSKL (Mathur et al., 2002) 492 and StCFP (Brandizzi et al., 2002) constructs were infiltrated at 0.04 optical density. 493 Leaf samples (~5mm2) were taken from plants after 3-4 days expression and 494 incubated in 25μM latrunculin b for 60 minutes prior to imaging. 495 496 Confocal imaging and determination of organelle association and residency time 497 with chloroplasts 498 Triple imaging of peroxisomes (YFPSKL), Golgi (StCFP) and chloroplasts 499 (autofluorescence) in live tobacco epidermal pavement cells was done using multi- 500 tracking in line switching mode on a Zeiss LSM510 Meta confocal microscope. CFP 501 was excited with a 458-nm argon laser and YFP/ chloroplast autofluorescence with a 502 514nm laser, their emissions passed through a HFT 458/514 main dichroic beam 503 splitter and NFT 490 and NFT 595 secondary dichroic beam splitter, and detected 504 using 470-500nm, 530-600nm and 647-690nm filters respectively. All imaging was 505 carried out using a 63 x 1.4 Numerical Aperture (NA) oil immersion objective with a 506 scan speed of 1.94 frames per second. Peroxisomes / Golgi which were up to 1μm 507 from the chloroplasts (as monitored by the autofluorescent signal) were categorised as 508 residing next to chloroplasts. Residency time of peroxisomes and Golgi on 509 chloroplasts were analysed manually. Only those which resided next too (and could 510 move laterally over the surface of) the chloroplast for more than 3 seconds were 511 included in the statistical analysis. 512 Dual imaging of peroxisomes (GFPSKL) and chloroplasts (autofluorescence) was 513 carried out using multi-tracking in line switching mode on a Zeiss LSM510 Meta 514 confocal microscope. GFP was excited with a 488-nm argon laser and 515 autofluorescence with a 514nm laser, their emissions passed through a HFT 488/543 516 main dichroic beam splitter and NFT 515 and NFT 545 secondary dichroic beam 517 splitter, and detected using 505-530nm and 636-690nm filters respectively. All 518 imaging was carried out using a 63 x 1.4 NA oil immersion objective. Twenty single 20 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 519 scans of a 143 x 143 μm area were taken, and the number of chloroplasts with a 520 juxtaposed peroxisome in each image was counted. 521 522 Optical trapping setup and data generation 523 Optical trapping was performed using a cw 1090 nm laser (SPI) focused using a x100, 524 oil immersion, NA 1.49 TIRF objective lens (Nikon). Here we assume the effective 525 NA of the objective lens for optical trapping approaches a value of 1.0. The 526 assumption is based upon comparison of escape force measurements made on 1.0 µm 527 diameter polystyrene beads to theoretical values calculated using an optical tweezers 528 computational toolbox (Nieminen et al., 2007). TIRF objectives are not commonly 529 used for optical tweezers. Mahamdeh et al. (2011) also indicate that spherical 530 aberrations arising from trapping in aqueous media will reduce the effective numerical 531 aperture. 532 TIRF used an excitation laser with 473 nm wavelength (Becker and Hickl) with a 533 maximum output power of 5 mW, coupled by an optical fibre to a Nikon TIRF 534 adapter system and attenuated by neutral density filters (2 and / or 8 dependent on the 535 level of GFP-SKL expression). Emitted fluorescent light was filtered using a long 536 pass filter for wavelength transmission above 505 nm and imaged using an electron- 537 multiplier charge-coupled device (Andor Ixon EMCCD). This allowed visualisation 538 of the excited GFPSKL probe and detection of chloroplast autofluorescence. Note, 539 that whilst the TIRF technique was employed to give significant improvement of 540 signal to noise, it is also likely that we are operating in a highly inclined illumination. 541 Custom LabVIEW® software (National Instruments) was used to control the EMCCD 542 camera (Andor), microscope stage (Marshauser) and a shutter, which blocked the 543 laser beam used for trapping. A LabVIEW® interface was used to synchronise the 544 timing of peroxisome capture, stage translation and peroxisome release over 110 or 545 229 frame videos; peroxisomes were monitored for 10 frames prior to trap activation, 546 40 frames upon trap activation prior to movement, 10 frames for the 6μm translation, 547 10 frames after the translation, and 40 or 159 frames after the trap was deactivated 548 (relating to 5.3second or 21.5second recovery periods respectively). Stage translation 549 was measured to be 5.74μm in 1 second with the EMCCD cycle time of 0.135 550 seconds giving approximately 7.5 frames per second. The video sequences were 551 stored as 16-bit stacked “tagged image file format” (tiff) files for subsequent analysis 21 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 552 of peroxisome behaviour. Note, the data sets generated for figure 2 are a combination 553 of the above trapping routine and an earlier version where trap shuttering was 554 manually controlled over a 70 frame video. 555 556 The minimal force (i.e. the escape force) required to trap peroxisomes, in a non-cp 557 environment, were measured by application of a viscous drag force (Supp Fig. 2). The 558 laser trap strength and viscous properties were investigated using a set of controlled 559 experiments where the stage velocity was varied. For each stage velocity, the laser 560 power required to keep 50% of the captured peroxisomes in the optical trap was 561 determined over a fixed 6μm translation distance (Supp Fig. 2). The fluorescent 562 organelles were observed under TIRF illumination. Due to variability in peroxisome 563 diameter it was necessary to measure 30-80 peroxisomes at each stage velocity to 564 obtain a representative laser power. Thus, the reported laser power is for an “average” 565 peroxisome (with plotted error bars indicating S.E uncertainty in laser power). The 566 viscous drag force for each stage velocity was calculated using Stokes’ law with an 567 assumed viscosity value of 0.06 Pa s (Scherp and Hasenstein, 2007) and the average 568 measured peroxisome diameter. Error bars for viscous drag force calculations used the 569 S.E. variation of peroxisome diameter. As a control, the same procedure was applied 570 to 1 μm diameter polystyrene beads in water (0.00089 Pa s). 571 572 Analysis of optical trapping data 573 Trapping data from each repetition was normalised against differences in sample size 574 to determine the percentages of peroxisomes that were either trapped, untrapped or 575 which escaped the trap per leaf sample. The weighted mean values were taken of 576 these percentages for whole datasets and plotted. Between 36 and 62 peroxisomes in 577 total underwent the trapping protocol at any given laser trapping power resulting in 578 n=338 for chloroplast and n=381 for non-chloroplast associated total sample sizes. 579 These totals represent between 5 and 9 repetitions, where each repetition is from 1 580 leaf sample taken from 6 – 9 independent plants. Trapping was only attempted once 581 per peroxisome, repeated trapping of the same peroxisome was not undertaken. 582 583 Displacement values for peroxisome and peroxule dynamics were carried out using 584 ImageJ (NIH). 22 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 585 586 In order to gather statistically significant peroxisome motion data, we developed a 587 customized detection and tracking algorithm using a combination of python (scipy) 588 and custom written scripts and algorithms. The data were first filtered using the 589 Laplace-of-Gaussian scale-space method (Lindenberg 1994) to selectively filter for 590 objects in a given size range. Next, robust image statistics based thresholding (Median 591 Absolute Deviation) selected only salient objects in the resulting filtered data as 592 outlined in Murtagh et al. (2000). Object tracking was performed using a Global 593 Nearest Neighbours point registration approach, implemented as a modified version of 594 the Jonker-Volgenant linear assignment problem algorithm, altered to allow 595 rectangular cost matrices and cost cut-offs. In addition, sub-pixel peroxisome 596 positions were calculated using a filtered intensity weighted centroid function. 597 Tracking validation was performed by manual verification. The resulting trajectories 598 were then analysed to determine the peroxisome motion between the moment that the 599 optical trap was disengaged and the end of the recovery period. 600 601 Force calculations are described in the spring model (see supplementary note). 602 603 604 Supplementary data 605 606 Supplementary Figure 1. Relationship between cp and non-cp peroxisomal 607 behaviour in the optical trap. 608 609 Supplementary Figure 2: Laser power required for trapping peroxisomes and 610 polystyrene beads at different stage velocities. 611 612 Supplementary Figure 3: Spring model definition. 613 614 Supplementary Figure 4: Example fits to the data using the simple spring model. 615 616 Supplementary Table 1: Relationship between optical laser trap power and 617 peroxule formation characteristics from cp and non-cp peroxisomes. 618 23 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 619 Supplementary Movie 1. Peroxisome association with chloroplasts 620 621 Supplementary Movie 2. Peroxisomes can be trapped and moved laterally within 622 tobacco leaf epidermal cells. 623 624 Supplementary Movie 3. Peroxisome behaviour in the optical trap under actin 625 depolymerisation. 626 627 Supplementary Note. Spring Model of Peroxisome Motion 628 629 Acknowledgements 630 IS and HBG are funded by the BBSRC (BB/I006184/2) and an STFC user access 631 facility grant (LSP907), JM is funded by a Wellcome Trust Institutional Strategic 632 Support Award (WT097835MF) and NAT by The Leverhulme Trust and STFC. We 633 would like to thank Anne Kearns and Ian Leaves for technical assistance with 634 growing the tobacco plants required for the experiments, and to members of the plant 635 cell biology group at Oxford Brookes University for help with experimental logistics. 636 YFPSKL and StCFP were gifts from Prof. J. Mathur and Prof. C. Hawes respectively. 637 638 Author contributions 639 Experiments were conceived by IS and experimental data generated by HBG and IS. 640 ADW, SWB, BC and MRP built, customised, maintained and facilitated the use of the 641 optical trap-TIRF system. Experimental design was discussed with HBG, ADW, 642 SWB and JM. ADW calibrated the system and performed bead trapping experiments. 643 HBG performed all Image J analysis. JM wrote the tracking algorithm to generate the 644 displacement values for NAT. Visual confirmation of tracking was carried out by JM, 645 HBG and IS. NAT applied the simple model of a viscously-damped sphere on a 646 spring to determine the forces involved in the system. IS wrote the manuscript with 647 comments from all authors. NAT wrote the supplementary note section. NAT and 648 ADW were involved in writing the section relating to forces. 649 650 Figure legends 651 24 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 652 Figure 1. Optical trapping and movement of peroxisomes away from 653 chloroplasts in tobacco leaf epidermal cells. 654 Schematic representation of the trapping procedure (A-D) and the corresponding 655 micrographs (E-H) are shown. Upon turning the trap on (A,E) and moving the stage 656 6μm at a set speed (B,F; referred to as translation period) the trapped peroxisome (p, 657 white arrow) is pulled away from the chloroplast (cp) and a peroxule (arrowhead) is 658 formed. Upon turning the trap off (C,G) the peroxisome recoils backs towards its 659 original position next to the chloroplast (D,H). Peroxisome displacement during the 660 recovery period (referred to as recovery displacement) is measured (double 661 arrowhead). Asterisk denotes the tip of the peroxule. Scale bar 6μm. 662 663 Figure 2. Higher optical trapping laser power is required to trap and move 664 peroxisomes away from chloroplasts. 665 Non-cp (A) and Cp (B) associated peroxisomes underwent the optical trapping 666 protocol using various trapping laser powers and their trapping characteristics were 667 scored; remained in the trap over the 6μm translation (black bar A,B), unable to be 668 trapped (white bar A,B) or escaped the trap during the translation (grey bar A,B). 669 Percentages displayed are based on weighted means from a set of independent 670 experiments. Supp Figure 1 compares cp with non-cp for all three trapping categories 671 and indicates significant differences between peroxisomes that are trapped or escape 672 from the trap for cp versus non-cp. Relationship between optical laser trap power and 673 peroxule formation are given in Supp table 1. 674 Stills from Supp movie 3C-E representing before and after translation events for 675 peroxisomes which are trapped, not trapped or escape the trap during the translation 676 event are displayed (C, arrowhead). Note, peroxisomes not subjected to trapping in 677 the same cell are shown for comparison (arrow). The translation event is based on 678 movement of the stage and not the trap. Composite image of frames captured during 679 the translation event show that the trapped peroxisome does not appear to move 680 whereas organelles that escape the trap or are not trapped result in comet like tails 681 (merged panels). Scale bar 6μm. 682 25 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 683 Figure 3. Correlation between peroxisome displacement during the recovery 684 period and peroxule tip displacement during translation indicates anchored 685 tethering between chloroplasts and peroxisomes. 686 Peroxule tip displacement during the translation period was plotted against the 687 peroxisome displacement during the recovery period for chloroplast associated (A; 688 n=37) and non associated peroxisomes (B; n=46). Peroxisomes were trapped with 689 37mW optical trap laser power followed by a 21.5 second recovery period. The 690 behaviour of cp samples is indicative of anchored tethers where the peroxule tip 691 represents the base of the tether: small peroxule tip displacement combined with large 692 peroxisome recovery displacement (circle). Note, the sample sizes are different to 693 those in supplementary table 1 as displacement could only be measured if the 694 peroxule was observable for the entire period. 695 696 Figure 4. Cumulative distribution functions of model parameters for cp and non- 697 cp associated peroxisomes. 698 Recovery displacement b is larger for cp than non-cp peroxisomes (A), whereas k/μ 699 values, indicative of the tether stiffness, are broadly similar (B). Also, shown are the 700 derived spring constants (C) and initial recovery forces (D) calculated assuming a 701 viscosity of 0.06 Pa s for the cytoplasm. Values are derived using the spring model 702 described in the supplementary information alongside Supp Fig 2-4. 703 704 Supplementary material 705 706 Supplementary Movie 1. Peroxisome association with chloroplasts. 707 Time lapse images were taken of peroxisomes (green), Golgi (cyan) and chloroplasts 708 (magenta) in tobacco leaf epidermal cells. Organelles were visualised through 709 transient expression of fluorescent fusions (YFPSKL for peroxisomes and STCFP for 710 Golgi bodies) or autofluorescence (chloroplasts). Compared to Golgi, peroxisomes 711 spend longer periods of time associated with chloroplasts. The peroxisome appears 712 tethered to a fixed zone on the surface of the chloroplast as the chloroplast moves (A), 713 and in some cases the peroxisomes can also move laterally over the surface (B). Scale 714 bar 5 μm. 26 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. 715 716 Supplementary Movie 2. Peroxisomes can be trapped and moved laterally within 717 tobacco leaf epidermal cells. 718 Peroxisomes were trapped (arrowhead) and the stage moved 6 μm horizontally. 719 During the translation peroxisomes either escaped the trap (A,C) or were moved 6 μm 720 (B,D). Upon turning the trap off the peroxisomes moved back towards their original 721 position (B,D). Peroxisomes juxtaposed to chloroplasts (C,D) behaved similarly to 722 peroxisomes which were not (A,B). In both cases, peroxules were observed (B,D) 723 Scale bar 6 μm. 724 725 Supplementary Movie 3. Peroxisome behaviour in the optical trap under actin 726 depolymerisation. 727 A trapped non-cp (A) and cp (B) peroxisome undergoes the 6 μm translation resulting 728 in peroxule formation. Upon turning the trap off the peroxisome moves back along the 729 length of the peroxule. Movies highlighting examples of non-cp peroxisomes in 730 tobacco leaf epidermal cells which are either (C) trapped, (D) not trapped (E) or 731 escape the trap over the translation period. The samples were treated with latrunculin 732 B and so any subsequent motion upon turning the trap off is independent of acto- 733 myosin. Scale Bar 6 μm. Arrow head denotes peroxisome undergoing the trapping 734 routine. 735 736 27 Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Parsed Citations Andersson, MX., Goksor, M., and Sandelius, AS. (2007) Optical manipulation reveals strong attracting forces at membrane contact sites between endoplasmic reticulum and chloroplasts. J. Biol. Chem. 282:1170-1174. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Asada, K. (2006) Production and scavenging of reactive oxygen species in chloroplasts and their functions. Plant Physiol. 141:391396. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Avisar, D., Prokhnevsky, AI., Makarova, KS., Koonin, EV., and Dolja, VV. (2008) Myosin XI-K Is required for rapid trafficking of Golgi stacks, peroxisomes, and mitochondria in leaf cells of Nicotiana benthamiana. Plant Physiol. 146:1098-1108. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Barton, K., Mathur, N., and Mathur, J. (2013) Simultaneous live-imaging of peroxisomes and the ER in plant cells suggests contiguity but no luminal continuity between the two organelles. Front. Physiol. 4:196. doi: 10.3389/fphys.2013.00196. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Brandizzi F., Snapp EL., Roberts AG., Lippincott-Schwartz J., Hawes C. (2002) Membrane protein transport between the endoplasmic reticulum and the Golgi in tobacco leaves is energy dependent but cytoskeleton independent. Plant Cell 14:12931309. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Buchnik, L., Abu-Abied, M., and Sadot, E. (2015) Role of plant myosins in motile organelles: Is a direct interaction required? Journal of integrative plant biology 57:23-30. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Bussell, J.D., Behrens, C., Ecke, W., and Eubel, H. (2013) Arabidopsis peroxisome proteomics. Front. Plant Sci. 4:101. doi: 10.3389/fpls.2013.00101 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Cutler, SR., Ehrhardt, DW., Griffitts, JS., and Somerville, CR. (2000) Random GFP::cDNA fusions enable visualization of subcellular structures in cells of Arabidopsis at a high frequency. PNAS 97:3718-3723. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Desai M, Hu J. (2008) Light induces peroxisome proliferation in Arabidopsis seedlings through the photoreceptor phytochrome A, the transcription factor HY5 HOMOLOG, and the peroxisomal protein PEROXIN11b. Plant Physiol. 146:1117-27. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Frederick, SE., and Newcomb, EH. (1969).Microbody-like organelles in leaf cells. Science 163:1353-1355 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Hammer, JA., 3rd, and Sellers, JR. (2012) Walking to work: roles for class V myosins as cargo transporters. Nat. Rev. Mol. Cell Biol. 13:13-26. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Hanson, MR., and Sattarzadeh, A.(2013).Trafficking of proteins through plastid stromules. Plant Cell 25: 2774-2782. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Higa T, Suetsugu N, Kong SG, Wada M. (2014) Actin-dependent plastid movement is required for motive force generation in directional nuclear movement in plants. PNAS 111:4327-31. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title from on June 17, Published by (2012) www.plantphysiol.org Hu, J., Baker, A., Bartel, B., Linka, N.,Downloaded Mullen, RT., Reumann, S.,2017 and-Zolman, BK. Plant peroxisomes: biogenesis and Copyright © 2015 American Society of Plant Biologists. All rights reserved. function. Plant Cell 24: 2279-2303. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Jedd, G., and Chua, NH. (2002) Visualization of peroxisomes in living plant cells reveals acto-myosin-dependent cytoplasmic streaming and peroxisome budding. Plant Cell Physiol. 43:384-392. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Knoblach, B., Sun, X., Coquelle, N., Fagarasanu, A., Poirier, RL., and Rachubinski, RA. (2013) An ER-peroxisome tether exerts peroxisome population control in yeast. EMBO J. 32: 2439-2453. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Lindeberg T. (1994) Scale-space theory in computer vision. eds Kluwer Academic Publishers, Boston. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Madison, SL., and Nebenführ, A. (2013). Understanding myosin functions in plants: are we there yet? Curr. Opin. Plant. Biol. 16:710717 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Mahamdeh M, Campos CP., Schäffer E. (2011) Under-filling trapping objectives optimizes the use of the available laser power in optical tweezers. Opt. Express 19:11759-11768. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Mano, S., Nakamori, C., Hayashi, M., Kato, A., Kondo, M., and Nishimura, M. (2002) Distribution and characterization of peroxisomes in Arabidopsis by visualization with GFP: dynamic morphology and actin-dependent movement. Plant Cell Physiol. 43:331-341. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Mathur, J., Barton KA., Schattat, MH. (2013) Fluorescent Protein Flow within Stromules Plant Cell 25: 2771-2772. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Mathur, J., Mammone, A., and Barton, KA. (2012) Organelle extensions in plant cells. J. Integr. Plant. Biol. 54, 851-867 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Mathur, J., Mathur, N., and Hulskamp, M. (2002) Simultaneous visualization of peroxisomes and cytoskeletal elements reveals actin and not microtubule-based peroxisome motility in plants. Plant Physiol. 128:1031-1045 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Mehrshahi, P., Stefano, G., Andaloro, JM., Brandizzi, F., Froehlich, JE., DellaPenna, D. (2013) Transorganellar complementation redefines the biochemical continuity of endoplasmic reticulum and chloroplasts. PNAS 110:12126-12131 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Murtagh F, Starck JL. (2000) Image processing through multiscale analysis and measurement noise modeling. Stat. Comput. 10:95103. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Nieminen TA., Loke VLY., Stilgoe AB., Knoner G., Branczyk AM. (2007) Optical tweezers computational toolbox. J. Opt. Pure Appl. Opt. 9:S196-S203. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Oikawa, K., Kasahara, M., Kiyosue, T., Kagawa, T., Suetsugu, N., Takahashi, F., Kanegae, T., Niwa, Y., Kadota, A., and Wada, M. (2003) Chloroplast unusual positioning1 is essential for proper chloroplast positioning. Plant Cell 15:2805-2815. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Oikawa, K., Matsunaga, S., Mano, S.,Downloaded Kondo, M.,from Yamada, K.,17, Hayashi, M., Kagawa, T., Kadota, A., Sakamoto, W., Higashi, S., on June 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Watanabe, M., Mitsui, T., Shigemasa, A., Iino, T., Hosokawa, Y., Nishimura, M. (2015) Physical interaction between peroxisome and chloroplasts elucidated by in situ laser analysis. Nature Plants 1:15035 doi:10.1038/nplants.2015.35 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Oikawa, K., Yamasato, A., Kong, S., Kasahara, M., Nakia, M., Takahashi, F., Ogura, Y., Kagawa, T., Wada, M. (2008) Chloroplast outer envelope protein CHUP1 is essential for chloroplast anchorage to the plasma membrane and chloroplast movement. Plant Physiology 148:829-842 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Prestele, J., Hierl, G., Scherling, C., Hetkamp, S., Schwechheimer, C., Isono, E., Weckwerth, W., Wanner, G., and Gietl, C. (2010) Different functions of the C3HC4 zinc RING finger peroxins PEX10, PEX2, and PEX12 in peroxisome formation and matrix protein import. PNAS 107:14915-14920 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Prinz, WA. (2014) Bridging the gap: membrane contact sites in signaling, metabolism, and organelle dynamics. J. Cell Biol. 205:759769. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Sandalio, L.M. and Romero-Puertas, M.C. (2015) Peroxisomes sense and respond to environmental cues by regulating ROS and RNS signalling networks. Annals of Botany 116:475-485 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Schattat, M., Barton, K., Baudisch, B., Klosgen, RB., and Mathur, J. (2011) Plastid stromule branching coincides with contiguous endoplasmic reticulum dynamics. Plant Physiol. 155:1667-1677. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Scherp, P., and Hasenstein, KH. (2007) Anisotropic viscosity of the Chara (Characeae) rhizoid cytoplasm. Am. J. Bot. 94:1930-1934. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Schmidt von Braun, S., and Schleiff, E. (2008) The chloroplast outer membrane protein CHUP1 interacts with actin and profilin. Planta 227:1151-1159 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Schnarrenberger, C., and Burkhard, C. (1977) In-vitro interaction between chloroplasts and peroxisomes as controlled by inorganic phosphate. Planta 134:109-114. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Schumann, U., Prestele, J., O'Geen, H., Brueggeman, R., Wanner, G., and Gietl, C. (2007) Requirement of the C3HC4 zinc RING finger of the Arabidopsis PEX10 for photorespiration and leaf peroxisome contact with chloroplasts. PNAS 104:1069-1074 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Scott, I., Sparkes, IA., and Logan, DC. (2007) The missing link: inter-organellar connections in mitochondria and peroxisomes? Trends Plant Sci. 12:380-381. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Sinclair, AM., Trobacher, CP., Mathur, N., Greenwood, JS., Mathur, J. (2009) Peroxule extension over ER-defined paths constitutes a rapid subcellular response to hydroxyl stress. Plant J. 59:231-242 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Sparkes, I (2011) Recent advances in understanding plant myosin function: Life in the fast lane. Mol Plant 4:805-812 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Sparkes IA., Brandizzi F, Slocombe SP, El-Shami M, Hawes C, Baker A. (2003) An Arabidopsis pex10 null mutant is embryo lethal, implicating peroxisomes in an essential role during plant embryogenesis. Plant Physiol. 133:1809-1819. Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Sparkes, IA., Ketelaar, T., de Ruijter, NC., and Hawes, C. (2009b) Grab a Golgi: laser trapping of Golgi bodies reveals in vivo interactions with the endoplasmic reticulum. Traffic 10:567-571. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Sparkes, IA., Teanby, NA., and Hawes, C. (2008) Truncated myosin XI tail fusions inhibit peroxisome, Golgi, and mitochondrial movement in tobacco leaf epidermal cells: a genetic tool for the next generation. J. Exp. Bot. 59:2499-2512. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Sparkes I., Runions, J., Hawes, C., Griffing L. (2009a) Movement and remodelling of the Endoplamic Reticulum in nondividing cell sof tobacco leaves. Plant Cell 21:3937-3949. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Sparkes IA., Runions J., Kearns A., Hawes C. (2006) Rapid, transient expression of fluorescent fusion proteins in tobacco plants and generation of stably transformed plants. Nat. Protoc. 1:2019-2025. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Stefano, G., Hawes, C., and Brandizzi, F. (2014) ER - the key to the highway. Curr. Opin. Plant Biol. 22:30-38. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Tamura K, Iwabuchi K, Fukao Y, Kondo M, Okamoto K, Ueda H, Nishimura M, Hara-Nishimura I. (2013) Myosin XI-i links the nuclear membrane to the cytoskeleton to control nuclear movement and shape in Arabidopsis. Curr. Biol. 23:1776-81. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Thazar-Poulot N, Miquel M, Fobis-Loisy I, Gaude T (2015) Peroxisome extensions deliver the Arabidopsis SDP1 lipase to oil bodies. PNAS 112:4158-63 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Yang, Y., Sage, TL., Liu, Y., Ahmad, TR., Marshall, WF., Shiu, SH., Froehlich, JE., Imre, KM., and Osteryoung, KW. (2011) CLUMPED CHLOROPLASTS 1 is required for plastid separation in Arabidopsis. PNAS 108:18530-18535. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Downloaded from on June 17, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved.